Building ventures, or helping the people who do.

The short version

What you get. A sourcing desk that runs from week one: a ranked digest of new rounds twice a week, a founder file and call questions for every company you want to meet, and a one-page memo within a day of asking.1
Why me. I've built a company on my own money, built the sourcing tool I'd use at your fund, and I put my calls on record so you can check them later.2 3
The ask. Eight weeks on site, from 4 January to 26 February 2027, at a fund or a venture-backed startup in the Gulf or Singapore.

50+early-stage startups screened since 2025

60GCC companies tracked, each checked against its source

20questions investors keep asking, drawn from 134 put to 44 founders

32forecasts on record in Signal Scout, scored as they settle

Before you read about me

Three companies I'd bring to your Monday meeting

From my scan of Gulf and India-linked rounds announced 31 October to 8 November 2026. Each one in five parts: the company, its numbers, why now, the biggest risk and a dated call you can score me on. Every number links to its source; company figures and my own arithmetic are marked.

mamun

Muscat, with entities in Abu Dhabi and Hong Kong

Pursue

$3M seed · 4 Nov 2026 · Phaze Ventures (lead, Oman Fund), Jasoor Ventures

Sharia-compliant trade finance with AI underwriting for the Greater Bay Area, Oman and UAE corridor. Licensed by Oman's Financial Services Authority (CMI051).

40%of settlements already clear through Hong Kong (company figure)

#26Rising Star, Deloitte Technology Fast 50 Middle East 2025

41%of SME trade-finance applications are turned down (ADB, cited in the release)

Why now: Oman's non-oil exports are up about 10% and re-exports 13.6% through July 2026 (company release), and mamun is already licensed and settling in three places.
The biggest risk: Credit losses on first-time importers with no history. No loss rate is public.

My call

25%Closes a Series A of $8M or more by 31 Dec 2028

Base rate 7 to 25%: MENA early-stage startups that ever reach Series A (MAGNiTT, 2026) to seed companies that do within two years (Carta, 2026). I sit at the top: licensed, settling in three jurisdictions, already ranked. Against it: MENA early-stage deals fell 42% in 9M 2026.

I cut it to 15% if no volume figure is public by June 2027.

What I'd check first: Loss rates by corridor, and how the model prices a first-time importer with no history.

Sources: Wamda · Media OutReach · MAGNiTT · Carta

Clinics On Cloud

Pune, with the UAE as its global hub

Track

₹120 Cr (about $12.5M) Series A at a ₹600 Cr (about $63M) valuation · 3 Nov 2026 · Model42 (Enlightened Minds Investments, UAE)

AI health ATMs, box clinics and mobile units for preventive screening. Founded 2018 and bootstrapped until this round.

3,500+deployments in 200 Indian cities and 8 countries (company figure)

12M+patients screened to date (company figure)

$18Kof valuation per deployment (my arithmetic: $63M / 3,500)

Why now: Bootstrapped for eight years, it now has ₹120 Cr and a UAE-based lead investor to take a proven Indian product abroad.
The biggest risk: UAE health-regulator approval takes months, and kiosk economics built for Indian prices may not hold at Gulf costs.

My call

60%Names a paying UAE customer or a live UAE site by 30 Sep 2027

No sourced base rate. For it: the UAE is the stated hub, the lead investor is UAE-based and co-founded by two doctors, and the cash is in hand. Against it: devices and clinics need UAE health-regulator approval, which takes months.

I cut it to 40% if no UAE partner is named by March 2027.

What I'd check first: Who pays per screening in the UAE (employers, insurers or patients), and whether kiosk economics built for Indian prices hold at Gulf costs.

Sources: Inc42 · StartupTalky

Persona

GCC-based venture studio

Track

$630K pre-seed · 6 Nov 2026 · Upturn (lead)

Builds consumer brands with creators and media owners, who share ownership. Eight brands across food and beverage, FMCG, beauty, retail and lifestyle (company figure).

8brands live across five categories (company figure)

$79Kof new capital per brand (my arithmetic: $630K / 8)

42%fall in MENA early-stage deals in 9M 2026 (MAGNiTT)

Why now: Gulf seed money is scarce (early-stage deals fell 42% in 9M 2026), which favours a studio that launches a brand on about $79K.
The biggest risk: Eight brands on $630K is thin, and a creator who leaves may take the audience with them.

My call

30%Raises a seed round of $2M or more by 31 Dec 2027

No sourced base rate for Gulf pre-seed to seed. Against it: seed money is scarcer and $79K a brand is thin. For it: eight live brands is real output for a pre-seed studio, and one breakout brand would carry the round.

I cut it to 20% if no brand-level revenue is shared by June 2027.

What I'd check first: Revenue concentration across the eight, repeat purchase per brand, and who owns the audience when a creator leaves.

Sources: MENA Startup Digest · The Startup Scene · MAGNiTT

Tell me your thesis, and I'll send three for your fund within 24 hours.Send me your thesis

§1What you get

Each of these starts in week one.

What I'd add to your fund

Five things, each one already working, so you can judge the output before you judge me.

Deal flow you'd otherwise miss

Rounds from the Gulf, Singapore and the India corridor, screened against your thesis, ranked and argued, every Sunday and Wednesday.

See Signal Scout's feed

A founder file before every first call

What the founders claim and whether a registry or a second source backs it, their own words, and the questions that matter most for that company.

See Clueso's file

A one-page memo within a day

Facts, estimates and predictions, each labelled, so a partner sees at a glance what's known and what's a bet.

See the Clueso memo

Calls you can audit

Every prediction I make carries a probability, a date and a test, and gets scored when it settles. By week eight you'll know how far to trust me.

See the track record

Operator help for your founders

I've built a Shopify storefront from scratch, run Meta ads, imported from China and sold into the US on US hours. Consumer and commerce founders can use that from day one.

Ask me in the call

§2How I think

The rules I'd hold a founder to. I hold myself to them first.

Five rules, and where each one came from

None of these came from a book. Each one came from something I built, sold or got wrong.

Find the rule that can kill a deal before negotiating anything else.

Where it came from: I found an enterprise logistics partnership with Samsung SDS on my own and took it to final commercial terms. It stopped at a vendor rule, a minimum operating history we didn't yet meet. I could have found that rule on day one.
In my work: Signal Scout runs a fund's hard filters before it scores anything.

Write down what you believe, with a number and a date.

Why: A forecast without a date can't be wrong, so it can't teach you anything.
In my work: Scout marks every line as a fact, an estimate or a prediction, starts each prediction from a base rate, and keeps score of its own calls.

Judge founders on behaviour, not pedigree.

Why: At the earliest stage the numbers are too young to decide anything, and pedigree is the market's shortcut. It misses operators whose edge is years spent inside a problem.
In my work: Founder Signal: the twenty questions investors ask, in order. What public data can't show becomes the agenda for the first call.

Treat capital efficiency as a skill, not a phase.

Where it came from: Violette's designs need laser-cut fabric and hardware that Indian makers mostly can't supply on time. I import, and I set up supply so almost no money sits in stock. The brand runs on about 30x less capital than a stock-led launch.
What I got wrong: I launched my first paid ads before purchase tracking was confirmed, so the test taught me less than it cost. Measure first, then spend.

Switch off your own system when it stops working.

Where it came from: I built OutreachOS on Google Sheets and Apps Script and seeded it across 650+ firms. It guessed email addresses. Most messages never arrived, and the ones that did were generic. I stopped using it.
In my work: This search runs on verified addresses, one message per person, written for that person.

§3Proof

Built in the last month. Open any of it.

What you can open today

Claims I can link, I've linked.

Product · October 2026

Signal Scout, a sourcing agent for venture funds

A fund sets its thesis: stage, round size, geography, sector or tech-agnostic, deep tech or not. Scout ranks new rounds against it, opens a founder file from the public record, and writes forecasts it can be held to. It reads founders from their own words, quotes its evidence and checks each quote against the source. It never scrapes LinkedIn or labels personality.

Thesis gate→Model tree→Founder file→Forecasts→Analyst vs red team→One-page memo

IPIDSingapore · Series A, $16M78Pursue

OneByZeroSingapore · Series A, $20M77Pursue

AilyticsSingapore · Series A, $7.4M77Pursue

1001Dubai · Series A, $30M69Track

From Scout's feed on 1 November 2026, under a growth-stage Asia thesis: the top three, and one it holds back because too little of its evidence is public yet. It runs a sourcing sweep every Sunday and Wednesday. Open Signal Scout

Founder Signal

Across 44 pitches the panel asked 134 questions. They were the same twenty, asked again and again in the same order.

The idea“What do they use today, and why would they switch to you?”

The proof“Do customers come back?”

The engine“Walk me through the money on one unit.”

The deal“Why this valuation?”

The founderDo they answer the question that was asked?

Research · 60 companies

Gulf Corridor Tracker

Early-stage companies across the GCC, each checked against its source and tagged when it has a verified India link. It's how I find the companies worth writing to funds about.

Memos · one page each

Pass or Pursue

Short screens that end in a call. The first covers Keep Converting, a UAE AI-commerce company, after its $2M pre-seed. The newest is Scout's first-look memo on Clueso.

§4Risks

A memo without risks is a pitch.

What could go wrong, and what I'd do about it

The four questions I'd ask if I were reading this about someone else.

Risk	What I'd do about it
01Eight weeks is short.	You'd see output in week one and a handover in week eight, so the work outlives my time with you.
02I haven't held a full-time fund role.	I've screened 50+ early-stage startups since 2025, and I built the tool I'd use at your desk. Judge the work before the title.
03I don't live in your city.	I'll be on site for all eight weeks, on the right pass for the role. Never a visit visa.
04I run a company of my own.	Violette has a small team around delivery, and my IIM Bangalore and final BTech terms both leave January and February free. Those eight weeks are yours.

§5Eight weeks

Output in week one, not week four.

What your eight weeks would look like

On site from 4 January to 26 February 2027. The work stays with your team when I leave.

Week 1

Learn how you decide

Your thesis and the last 20 companies you backed, passed on or missed. I set Signal Scout to your thesis and report how often it agrees with your 20 calls.

Weeks 2 to 6

Source and screen

Ten ranked digests, two a week, about 150 companies screened against your thesis, a founder file for every company you want to meet, and any memo you ask for within 24 hours.

Weeks 7 and 8

Keep score, hand over

Every call I made, scored: how many came true at each probability I gave. Then a handover so your team keeps the desk running.

If something else matters more to your team in January, I'll do that instead.

§6Background

Where I've been

Violette · Founder

Since 2024

A women's fashion brand, self-funded, sold online at violettefashion.in. Four collections and about 50 styles at ₹1,890 to ₹3,990, on a Shopify storefront I built myself. Shortlisted for the final interview of ISB's I-Venture Immersive Programme.

Lumens Depot · Business development

Since Oct 2026

Selling LED lighting into the US market, on US hours.

Independent startup evaluator and GTM adviser

Since 2025

Screened 50+ early-stage startups and advised some of their founders on go-to-market.

IIM Bangalore Entrepreneurship and Startup Wing · Core Coordinator

Since Nov 2026

Selected to lead the BBA DBE entrepreneurship and startup team.

Education

IIM Bangalore · BBA, Digital Business and Entrepreneurship

Class of 2028

Programmes

2025 and 2026

McKinsey.org Forward Program (Dec 2025) · Aspire Leaders Program (Jul 2026) · PPGF's ICPC political consulting internship (Oct 2026) · Finance and Operations Lead, AWS Student Builder Group

Projects

AI credit scoring for small businesses

Due Mar 2027

My final-year BTech computer science project: scoring small businesses from their UPI and GST sales.

FraudLens

2026

A UPI fraud-detection dashboard with explainable AI, built in a team of four for my BTech computer science course.

§7Contact

The next step is a call.

If this is worth a first call, tell me which company you're looking at. I'll send a one-page memo on it before we speak.

Emailarnav.bansal25@iimb.ac.in

LinkedInlinkedin.com/in/arnavbansall

Brandviolettefashion.in

ReadThe two-page memo · The eight-week plan · Research brief: the firms that last · CV

Based inDelhi NCR, India